Demultiplexing of S01Ven and S01control, cell annotation and CITE validation
David Mentrup
1 Preparations and Data
1.0 Required Packages
library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, unionlibrary(Seurat)
library(ggplot2)
library(scales)
library(mixtools)mixtools package, version 1.2.0, Released 2020-02-05
This package is based upon work supported by the National Science Foundation under Grant No. SES-0518772.library(mclust)Package 'mclust' version 5.4.6
Type 'citation("mclust")' for citing this R package in publications.
Attaching package: 'mclust'The following object is masked from 'package:mixtools':
dmvnormpackageVersion("dplyr")[1] '0.8.99.9003'packageVersion("Seurat")[1] '3.1.5'Pat1_HL60 <- readRDS(file ="./Pat1_HL60_CITE.rds")
DefaultAssay(Pat1_HL60 )<- "RNA"2 Remove Cellline and cells similiar to cell line
VlnPlot(Pat1_HL60, assay= "HTO",features = c("Pat1-R1-H1","Pat1-Ven-R1-H2", "HL60-R1-H3"), ncol = 2, pt.size = 0.1, group.by = "RNA_cluster") + NoLegend()
DimPlot(Pat1_HL60, label = TRUE, label.size = 10, reduction = "umap", group.by = "RNA_cluster") + NoLegend()Warning: Using `as.character()` on a quosure is deprecated as of rlang 0.3.0.
Please use `as_label()` or `as_name()` instead.
This warning is displayed once per session.
Pat1_CITE.c <- subset(Pat1_HL60, subset = `hto_HL60-R1-H3` < 1.5)
Pat1_CITE.c <- subset(Pat1_CITE.c, subset = RNA_cluster != 4 & RNA_cluster != 6 & RNA_cluster != 9)
VlnPlot(Pat1_CITE.c, assay= "HTO",features = c("HL60-R1-H3"), ncol = 1, pt.size = 0.1, group.by = "RNA_cluster") + NoLegend()
Pat1_CITE.cAn object of class Seurat
19524 features across 5364 samples within 3 assays
Active assay: RNA (19510 features, 2000 variable features)
2 other assays present: HTO, Epi
5 dimensional reductions calculated: pca, umap, pca_epi, tsne_epi, umap_epiDemultiplexing S01_CITE and S01_CITE_Ven
FeatureScatter(Pat1_CITE.c, feature1 = "hto_Pat1-R1-H1", feature2 = "hto_Pat1-Ven-R1-H2") + stat_function(fun=function(x)(1/4* exp (2*x)) -0.2 ) + stat_function(fun=function(x)(1/2*log (4*x) ) ) + ylim(0,4) + xlim(0,4)Warning: Removed 23 rows containing missing values (geom_point).Warning: Removed 65 row(s) containing missing values (geom_path).
Pat1_CITE.c[["H1.H2"]]<- Pat1_CITE.c@assays$HTO@data["Pat1-R1-H1",]-Pat1_CITE.c@assays$HTO@data["Pat1-Ven-R1-H2",]
VlnPlot(Pat1_CITE.c, features = c("H1.H2"), group.by = "orig.ident")+ geom_hline(yintercept=0.2) 
mixmdl <- Mclust(t(Pat1_CITE.c@assays$HTO@data[c("Pat1-R1-H1","Pat1-Ven-R1-H2"),]), G = 2)
summary(mixmdl, parameters = TRUE)----------------------------------------------------
Gaussian finite mixture model fitted by EM algorithm
----------------------------------------------------
Mclust VVV (ellipsoidal, varying volume, shape, and orientation) model with 2
components:
log-likelihood n df BIC ICL
-6826.522 5364 11 -13747.51 -14305.95
Clustering table:
1 2
2781 2583
Mixing probabilities:
1 2
0.4932864 0.5067136
Means:
[,1] [,2]
Pat1-R1-H1 0.4613017 1.2148829
Pat1-Ven-R1-H2 0.9045152 0.8719508
Variances:
[,,1]
Pat1-R1-H1 Pat1-Ven-R1-H2
Pat1-R1-H1 0.008072300 0.008268123
Pat1-Ven-R1-H2 0.008268123 0.060042503
[,,2]
Pat1-R1-H1 Pat1-Ven-R1-H2
Pat1-R1-H1 0.6492458 -0.1631349
Pat1-Ven-R1-H2 -0.1631349 0.7292268plot(mixmdl, what= "classification") 
mixmd2 <- normalmixEM(Pat1_CITE.c[["H1.H2"]][,1], k = 2)number of iterations= 26 data.frame(x = mixmd2$x)%>%
ggplot() +
geom_histogram(aes(x, ..density..), binwidth = 0.01, colour = "black",
fill = "white") + ylab("Density") +geom_vline (xintercept = 0.15)
post.df <- as.data.frame(cbind(x = mixmd2$x, mixmd2$posterior))
head(post.df, 100) x comp.1 comp.2
1 -0.24952714 7.475855e-01 0.2524145
2 0.59325213 2.119813e-06 0.9999979
3 -0.40265809 8.233979e-01 0.1766021
4 -0.50418790 8.147899e-01 0.1852101
5 -0.14986917 6.084293e-01 0.3915707
6 -0.16001123 6.269059e-01 0.3730941
7 -2.49369689 1.163713e-25 1.0000000
8 -1.64966567 5.574524e-09 1.0000000
9 -0.71390649 6.118769e-01 0.3881231
10 -0.73381256 5.730588e-01 0.4269412
11 -0.81303032 3.882069e-01 0.6117931
12 -0.79040474 4.444555e-01 0.5555445
13 -0.36868700 8.168388e-01 0.1831612
14 -1.08679220 1.230332e-02 0.9876967
15 -0.96304118 8.698854e-02 0.9130115
16 -0.33971683 8.071879e-01 0.1928121
17 -0.59364587 7.672856e-01 0.2327144
18 -0.50978874 8.130428e-01 0.1869572
19 0.90928691 6.512836e-11 1.0000000
20 0.08697600 9.985802e-02 0.9001420
21 -0.27071576 7.662715e-01 0.2337285
22 -0.48434460 8.198451e-01 0.1801549
23 -0.18890898 6.739850e-01 0.3260150
24 -0.97246964 7.649035e-02 0.9235097
25 -0.13555191 5.806367e-01 0.4193633
26 0.73517230 2.811048e-08 1.0000000
27 -1.23168867 6.696398e-04 0.9993304
28 3.07687788 2.538314e-74 1.0000000
29 0.54185450 8.837189e-06 0.9999912
30 0.53817774 9.759920e-06 0.9999902
31 0.63849572 5.677452e-07 0.9999994
32 -0.61619470 7.474940e-01 0.2525060
33 -0.03892519 3.539367e-01 0.6460633
34 0.39369948 3.590270e-04 0.9996410
35 0.71850051 4.808457e-08 1.0000000
36 0.74414945 2.098669e-08 1.0000000
37 0.59195925 2.199300e-06 0.9999978
38 -0.79834762 4.248635e-01 0.5751365
39 0.46850599 5.969536e-05 0.9999403
40 -0.57938918 7.779729e-01 0.2220271
41 0.83223184 1.059341e-09 1.0000000
42 -0.25979058 7.570450e-01 0.2429550
43 -0.12203155 5.526181e-01 0.4473819
44 -0.43974670 8.251299e-01 0.1748701
45 -0.43351386 8.252252e-01 0.1747748
46 0.29511821 3.005476e-03 0.9969945
47 -2.30989394 2.691358e-21 1.0000000
48 -0.29203857 7.819446e-01 0.2180554
49 0.90529214 7.556585e-11 1.0000000
50 -0.52082631 8.091694e-01 0.1908306
51 0.91448284 5.364338e-11 1.0000000
52 -1.39321218 1.296557e-05 0.9999870
53 -0.78029919 4.689952e-01 0.5310048
54 -0.64220372 7.197993e-01 0.2802007
55 3.68931554 1.671542e-102 1.0000000
56 -0.26404670 7.607402e-01 0.2392598
57 -0.25968834 7.569546e-01 0.2430454
58 -0.36850507 8.167902e-01 0.1832098
59 -0.55599225 7.927389e-01 0.2072611
60 -0.30035986 7.872787e-01 0.2127213
61 1.65271718 2.816214e-26 1.0000000
62 -0.50825781 8.135347e-01 0.1864653
63 0.94865248 1.470047e-11 1.0000000
64 2.57670077 1.161195e-54 1.0000000
65 -0.76773604 4.987111e-01 0.5012889
66 -0.58841309 7.713651e-01 0.2286349
67 -0.85755133 2.784040e-01 0.7215960
68 -0.91444232 1.586555e-01 0.8413445
69 0.87903906 1.985122e-10 1.0000000
70 1.05345525 2.265797e-13 1.0000000
71 -0.29007757 7.806257e-01 0.2193743
72 0.04425612 1.669251e-01 0.8330749
73 -0.94428715 1.110808e-01 0.8889192
74 -1.00120101 5.054634e-02 0.9494537
75 -1.91594722 2.578353e-13 1.0000000
76 -0.21591599 7.107519e-01 0.2892481
77 -0.20509708 6.968321e-01 0.3031679
78 -0.40760658 8.239503e-01 0.1760497
79 0.44817516 9.871955e-05 0.9999013
80 1.22190565 1.462412e-16 1.0000000
81 -0.51766649 8.103376e-01 0.1896624
82 -0.35057567 8.112683e-01 0.1887317
83 -0.39723627 8.226773e-01 0.1773227
84 -0.39713676 8.226629e-01 0.1773371
85 -0.80941256 3.972723e-01 0.6027277
86 -0.53143059 8.048906e-01 0.1951094
87 0.28377269 3.770460e-03 0.9962295
88 0.99580464 2.335763e-12 1.0000000
89 -0.47953081 8.208129e-01 0.1791871
90 -0.51188099 8.123529e-01 0.1876471
91 1.75929763 5.021009e-29 1.0000000
92 -0.30156262 7.880153e-01 0.2119847
93 -0.44995764 8.246369e-01 0.1753631
94 0.57767845 3.292526e-06 0.9999967
95 -0.40725104 8.239139e-01 0.1760861
96 3.57588399 6.080954e-97 1.0000000
97 -0.24230343 7.404462e-01 0.2595538
98 -0.21333731 7.075297e-01 0.2924703
99 -0.17787464 6.569758e-01 0.3430242
100 2.62927947 1.380705e-56 1.0000000GMM failed due to noise, hard borders selected based on kernel and overall expression
DefaultAssay(object = Pat1_CITE.c) <- "HTO"
Pat1_CITE.c@meta.data$Sample <- 'Unknown'
Pat1_CITE.c@meta.data[WhichCells(Pat1_CITE.c, expression = H1.H2 > 0.2) ,"Sample"] <- "Pat1_R1_H1-N"
Pat1_CITE.c@meta.data[WhichCells(Pat1_CITE.c, expression = H1.H2 < 0) ,"Sample"] <- "Pat1_R1_Ven_H2-N"
Pat1_CITE.c@meta.data[WhichCells(Pat1_CITE.c, expression = `Pat1-Ven-R1-H2` >1 & `Pat1-R1-H1` >1) ,"Sample"] <- "Multi-Cell-Hashs"
Pat1_CITE.c@meta.data[WhichCells(Pat1_CITE.c, expression = `Pat1-Ven-R1-H2` <0.2 & `Pat1-R1-H1` <0.2) ,"Sample"] <- "Unlabelled"
DefaultAssay(Pat1_CITE.c) <- "RNA"FeatureScatter(Pat1_CITE.c, feature1 = "hto_Pat1-R1-H1", feature2 = "hto_Pat1-Ven-R1-H2", group.by = "Sample") 
prop.table(table(Pat1_CITE.c@meta.data$Sample, Pat1_CITE.c@meta.data$seurat_clusters), margin = 1)*100
0 1 2 3 4
Multi-Cell-Hashs 7.3684211 11.0526316 57.8947368 17.8947368 0.0000000
Pat1_R1_H1-N 32.3782235 32.2063037 13.4097421 11.5759312 4.0687679
Pat1_R1_Ven_H2-N 39.5142180 34.5082938 10.6635071 8.0272512 2.5177725
Unknown 37.5000000 22.9166667 2.0833333 8.3333333 18.7500000
Unlabelled 60.0000000 20.0000000 0.0000000 20.0000000 0.0000000
5 6 7
Multi-Cell-Hashs 3.6842105 1.0526316 1.0526316
Pat1_R1_H1-N 1.9484241 2.1203438 2.2922636
Pat1_R1_Ven_H2-N 1.9549763 1.8957346 0.9182464
Unknown 6.2500000 4.1666667 0.0000000
Unlabelled 0.0000000 0.0000000 0.0000000prop.table(table( Pat1_CITE.c@meta.data$seurat_clusters, Pat1_CITE.c@meta.data$Sample), margin = 1)*100
Multi-Cell-Hashs Pat1_R1_H1-N Pat1_R1_Ven_H2-N Unknown Unlabelled
0 0.72388831 29.21406412 68.97621510 0.93071355 0.15511892
1 1.19318182 31.93181818 66.19318182 0.62500000 0.05681818
2 15.60283688 33.19148936 51.06382979 0.14184397 0.00000000
3 6.64062500 39.45312500 52.92968750 0.78125000 0.19531250
4 0.00000000 43.03030303 51.51515152 5.45454545 0.00000000
5 6.36363636 30.90909091 60.00000000 2.72727273 0.00000000
6 1.90476190 35.23809524 60.95238095 1.90476190 0.00000000
7 2.73972603 54.79452055 42.46575342 0.00000000 0.00000000table(Pat1_CITE.c@meta.data$Sample)
Multi-Cell-Hashs Pat1_R1_H1-N Pat1_R1_Ven_H2-N Unknown
190 1745 3376 48
Unlabelled
5 Reprocessing purified dataset (Just Pat1)
DefaultAssay(Pat1_CITE.c) <- "RNA"
Pat1_CITE.c<- NormalizeData(Pat1_CITE.c)
Pat1_CITE.c<- FindVariableFeatures(Pat1_CITE.c)
Pat1_CITE.c<- ScaleData(Pat1_CITE.c)Centering and scaling data matrixPat1_CITE.c<- RunPCA(Pat1_CITE.c)PC_ 1
Positive: SLC7A11, THBS1, FTH1, ANXA5, C15orf48, FCER1G, MS4A7, EPB41L3, NEAT1, CD14
CYBB, MMP14, PLA2G7, PLAUR, SH3BP5, CD93, S100A9, SOD2, KYNU, PILRA
IER3, SERPINB2, HNMT, CTB-61M7.2, MPEG1, CXCL16, VCAN, NCF2, S100A10, MAFB
Negative: RPL35, RPS2, NPM1, PTMA, LDHB, NME1, HSPE1, HSPD1, NCL, SRM
RANBP1, SNRPD1, NHP2, DUT, C1QBP, RAN, PA2G4, DCTPP1, PAICS, TYMS
PRSS57, SERBP1, PPA1, HSP90AB1, KIAA0101, GGCT, HNRNPA1, TUBA1B, SNRPE, PRKDC
PC_ 2
Positive: ZFAS1, MALAT1, TSC22D1, TFPI, EPB41L4A-AS1, IL11, DDIT3, NEU4, SNHG12, DDIT4
PPFIBP1, INHBA, BTG1, MEF2C, DDAH2, CLK1, SQSTM1, H3F3B, MTSS1, ZEB1
KDM6B, SMAD3, PPP1R15A, RSRC2, NR4A3, CREBRF, CHST7, RBKS, PHLDA1, USP53
Negative: ACTB, PFN1, TIMP1, CYP1B1, FCER1G, LGALS1, SLC44A1, COTL1, DNMT1, QSOX1
LST1, ALOX5AP, TYROBP, HSPA8, CAPG, PLA2G7, EPB41L3, ARPC5, TPI1, NCF2
RAB31, ITGB2, CD93, CYBB, NCF1, AIF1, SRGN, S100A9, MPEG1, CD14
PC_ 3
Positive: PHLDA1, INHBA, IL11, CXCL3, NAMPT, NEU4, TSC22D1, CXCL2, ATP2B1, PPP1R15A
BTG3, SAT1, H3F3B, DDIT3, RSRC2, IL8, BTG1, HMGA1, CTNNAL1, FOSL2
JUN, CXCL1, IL24, SQSTM1, KDM6B, LINC00936, NRIP3, DUSP1, CHST7, ATF3
Negative: PYCARD, ARHGDIB, CD74, HLA-DRB1, HLA-DRA, HLA-DMA, GSN, HLA-DPB1, AMICA1, TYROBP
S100A4, ACTB, TMSB10, HLA-DPA1, DOK2, HLA-DQB1, SORL1, C10orf54, LST1, CST3
PFN1, HCST, S100A6, HLA-DMB, CD52, CKLF, LGALS1, ENG, AIF1, LTB
PC_ 4
Positive: GIMAP7, CD3D, CD2, CXCR4, SPOCK2, GIMAP4, CD3E, CD7, CD3G, PIK3IP1
ETS1, SYNE2, CD247, CD48, LCK, GIMAP1, SELL, CAMK4, SESN3, PRDX2
CLEC2D, GZMM, RARRES3, IL7R, PASK, TMSB4X, RORA, CCR7, ARL4C, PDCD4
Negative: PPFIBP1, MEF2C, TXN, FTL, TNFRSF4, SERPINB1, PRDX1, SQSTM1, HSP90AB1, IL8
TFPI, HIST2H2AA3, CKLF, HMGA1, HLA-DRA, AGPAT9, HLA-DRB1, H2AFY, SAT1, AHR
ZFAS1, PPP1R15A, LMNA, MIR155HG, TPM4, MTSS1, SQLE, CD74, NFKB1, GSN
PC_ 5
Positive: IL6, CCL20, CXCL5, IL1A, CCL4, CXCL1, GATA2, PTGS2, IL1B, CCL3
RNF144B, MARCKS, MT1M, AQP9, G0S2, MT1G, MT2A, PPBP, TNFAIP6, TNIP3
SERPINB2, EREG, RP11-701P16.5, C1QTNF1, PRSS57, SLC7A7, AC003092.1, MT1E, MET, IL1R1
Negative: FGL2, LMNA, SAMHD1, TSC22D1, IL11, CD180, CTNNAL1, AGPAT9, PAK1, CKLF
FTL, CD300LB, DDIT3, ENPP2, RUNX3, RGCC, SLC8A1, FPR3, STK17B, NRIP3
RSRC2, C10orf54, CD36, JUN, NQO1, CCR1, H2AFZ, HMGA1, PKIB, PTPN22 ElbowPlot(Pat1_CITE.c, ndims = 50)
Only sample
Pat1_CITE.only <- subset(Pat1_CITE.c, subset = Sample == "Pat1_R1_H1-N" | Sample == "Pat1_R1_Ven_H2-N")
table(Pat1_CITE.only@meta.data$Sample)
Pat1_R1_H1-N Pat1_R1_Ven_H2-N
1745 3376 DefaultAssay(Pat1_CITE.only) <- "RNA"
Pat1_CITE.only<- NormalizeData(Pat1_CITE.only)
Pat1_CITE.only<- FindVariableFeatures(Pat1_CITE.only)
Pat1_CITE.only<- ScaleData(Pat1_CITE.only)Centering and scaling data matrixPat1_CITE.only<- RunPCA(Pat1_CITE.only)PC_ 1
Positive: RPL35, RPS2, NPM1, PTMA, RPL12, NME1, HSPE1, LDHB, HSPD1, NCL
SRM, RANBP1, SNRPD1, DUT, NHP2, C1QBP, RAN, TYMS, PA2G4, DCTPP1
PAICS, PRSS57, SERBP1, KIAA0101, PPA1, HSP90AB1, TUBA1B, GGCT, HNRNPA1, PRKDC
Negative: SLC7A11, THBS1, FTH1, ANXA5, C15orf48, NEAT1, MS4A7, FCER1G, EPB41L3, CD14
CYBB, MMP14, PLA2G7, PLAUR, SH3BP5, KYNU, IER3, SOD2, CD93, S100A9
PILRA, SERPINB2, CTB-61M7.2, CXCL16, HNMT, MPEG1, VCAN, NCF2, IL8, MAFB
PC_ 2
Positive: ACTB, PFN1, TIMP1, FCER1G, CYP1B1, LGALS1, SLC44A1, COTL1, ALOX5AP, PLA2G7
QSOX1, EPB41L3, CAPG, TYROBP, LST1, DNMT1, ARPC5, NCF2, CD93, HSPA8
RAB31, CYBB, TPI1, ITGB2, NCF1, S100A9, AIF1, CD14, MPEG1, MMP14
Negative: ZFAS1, MALAT1, TSC22D1, TFPI, EPB41L4A-AS1, IL11, DDIT3, NEU4, SNHG12, DDIT4
PPFIBP1, INHBA, MEF2C, BTG1, DDAH2, CLK1, MTSS1, SQSTM1, ZEB1, H3F3B
SMAD3, KDM6B, PPP1R15A, NR4A3, RSRC2, CREBRF, CHST7, AKIRIN2, RBKS, PHLDA1
PC_ 3
Positive: PYCARD, ARHGDIB, CD74, HLA-DRB1, HLA-DRA, HLA-DMA, HLA-DPB1, GSN, TYROBP, AMICA1
S100A4, TMSB10, ACTB, HLA-DPA1, DOK2, SORL1, HLA-DQB1, CD52, C10orf54, LST1
HCST, CST3, S100A6, HLA-DMB, PFN1, CKLF, LTB, AIF1, LGALS1, ENG
Negative: PHLDA1, IL11, INHBA, CXCL3, NEU4, TSC22D1, NAMPT, CXCL2, PPP1R15A, ATP2B1
SAT1, BTG3, H3F3B, DDIT3, RSRC2, IL8, HMGA1, BTG1, CTNNAL1, FOSL2
JUN, SQSTM1, KDM6B, IL24, CXCL1, LINC00936, NRIP3, DUSP1, CHST7, ATF3
PC_ 4
Positive: GIMAP7, CD3D, CD2, CXCR4, SPOCK2, GIMAP4, CD3E, CD3G, CD7, PIK3IP1
SYNE2, ETS1, CD247, CD48, LCK, GIMAP1, CAMK4, SELL, PRDX2, SESN3
GZMM, RARRES3, PASK, IL7R, TMSB4X, RORA, ARL4C, SLFN5, CLIC3, CCR7
Negative: PPFIBP1, MEF2C, TXN, TNFRSF4, FTL, SERPINB1, PRDX1, HSP90AB1, SQSTM1, CKLF
IL8, HLA-DRA, HLA-DRB1, TFPI, AGPAT9, HIST2H2AA3, HMGA1, TPM4, AHR, GSN
LMNA, ZFAS1, CD74, HLA-DMA, SAT1, SQLE, PFN1, PYCARD, MIR155HG, SYNJ2
PC_ 5
Positive: FGL2, LMNA, SAMHD1, TSC22D1, IL11, CD180, CKLF, PAK1, AGPAT9, CTNNAL1
RUNX3, CD300LB, ENPP2, SLC8A1, FTL, RGCC, DDIT3, C10orf54, FPR3, STK17B
CCR1, NQO1, NRIP3, IFI16, JUN, CD36, H2AFZ, PTPN22, RSRC2, GLIPR1
Negative: IL6, CCL20, CXCL5, IL1A, CCL4, GATA2, CXCL1, IL1B, CCL3, PTGS2
RNF144B, MT1M, MARCKS, AQP9, G0S2, MT1G, MT2A, TNFAIP6, TNIP3, PPBP
SERPINB2, EREG, RP11-701P16.5, C1QTNF1, PRSS57, MET, AC003092.1, IL1R1, MT1E, SLC7A7 ElbowPlot(Pat1_CITE.only, ndims = 50)
Pat1_CITE.only <- RunUMAP(Pat1_CITE.only, dims = 1:25)Warning: The default method for RunUMAP has changed from calling Python UMAP via reticulate to the R-native UWOT using the cosine metric
To use Python UMAP via reticulate, set umap.method to 'umap-learn' and metric to 'correlation'
This message will be shown once per session13:36:09 UMAP embedding parameters a = 0.9922 b = 1.11213:36:09 Read 5121 rows and found 25 numeric columns13:36:09 Using Annoy for neighbor search, n_neighbors = 3013:36:09 Building Annoy index with metric = cosine, n_trees = 500% 10 20 30 40 50 60 70 80 90 100%[----|----|----|----|----|----|----|----|----|----|**************************************************|
13:36:11 Writing NN index file to temp file /tmp/RtmpnfykTG/file28f9c2c8846cb
13:36:11 Searching Annoy index using 1 thread, search_k = 3000
13:36:12 Annoy recall = 100%
13:36:13 Commencing smooth kNN distance calibration using 1 thread
13:36:13 Initializing from normalized Laplacian + noise
13:36:14 Commencing optimization for 500 epochs, with 208674 positive edges
13:36:33 Optimization finishedPat1_CITE.only <- RunTSNE(Pat1_CITE.only, dims = 1:25)
Pat1_CITE.only <- FindNeighbors(Pat1_CITE.only, dims = 1:25)Computing nearest neighbor graph
Computing SNNPat1_CITE.only <- FindClusters(Pat1_CITE.only, resolution = 0.1,algorithm = 3, random.seed = 19950927)Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
Number of nodes: 5121
Number of edges: 179877
Running smart local moving algorithm...
Maximum modularity in 10 random starts: 0.9552
Number of communities: 6
Elapsed time: 3 secondsPat1_CITE.only@meta.data$RNA_cluster_0.1 <- Pat1_CITE.only@active.identprop.table(table(Pat1_CITE.only@meta.data$Sample, Pat1_CITE.only@meta.data$RNA_cluster_0.1), margin = 1)*100
0 1 2 3 4
Pat1_R1_H1-N 39.1977077 28.9398281 10.3724928 13.5816619 7.4498567
Pat1_R1_Ven_H2-N 47.5710900 23.9040284 13.0331754 9.1528436 5.7464455
5
Pat1_R1_H1-N 0.4584527
Pat1_R1_Ven_H2-N 0.5924171DimPlot(Pat1_CITE.only, reduction = "tsne", group.by="RNA_cluster_0.1", label = TRUE, label.size=10, pt.size = 0.5) + NoLegend()
DimPlot(Pat1_CITE.only, reduction = "umap", group.by="RNA_cluster_0.1", label = TRUE, label.size=10, pt.size = 0.5) + NoLegend()
Pat1_CITE.only <- FindClusters(Pat1_CITE.only, resolution = 0.5,algorithm = 3, random.seed = 19950927)Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
Number of nodes: 5121
Number of edges: 179877
Running smart local moving algorithm...
Maximum modularity in 10 random starts: 0.8741
Number of communities: 10
Elapsed time: 2 secondsPat1_CITE.only@meta.data$RNA_cluster <- Pat1_CITE.only@active.identprop.table(table(Pat1_CITE.only@meta.data$Sample, Pat1_CITE.only@meta.data$RNA_cluster), margin = 1)*100
0 1 2 3 4
Pat1_R1_H1-N 20.2292264 25.7306590 9.8567335 10.4297994 13.4670487
Pat1_R1_Ven_H2-N 24.8518957 21.7120853 16.3803318 13.1220379 9.0639810
5 6 7 8 9
Pat1_R1_H1-N 10.2578797 6.0171920 2.1203438 1.4326648 0.4584527
Pat1_R1_Ven_H2-N 7.4348341 4.5023697 1.0959716 1.2440758 0.5924171DimPlot(Pat1_CITE.only, reduction = "tsne", group.by="RNA_cluster", label = TRUE, label.size=10, pt.size = 0.5) + NoLegend()
DimPlot(Pat1_CITE.only, reduction = "umap", group.by="RNA_cluster", label = TRUE, label.size=10, pt.size = 0.5) + NoLegend()
Annotation with ConbAC
new.cluster.ids <- c("Leukemia cells (HSPC-alike)", "Leukemia cells (Monocytic progenitor-alike)", "Leukemia cells (stem cells)", "Leukemia cells (diffuse – Lymphocytes/Megakaryocyte patterns)", "Monocytes (and cDCs)", "Leukemia cells (Monocytic alike)", "Lymphocytes (CD4 T-Cell dominant)","Leukemia cells (stem cells 2)", "NK cells (and T-cells)", "Unknown (cDC dominant)")
names(new.cluster.ids) <- levels(Pat1_CITE.only)
Pat1_CITE.only <- RenameIdents(Pat1_CITE.only, new.cluster.ids)
Pat1_CITE.only@meta.data$celltype <- Pat1_CITE.only@active.ident
new.cluster.ids <- c("Leukemia (HSPC)", "Leukemia (Mon prog)", "Leukemia (stem)", "Leukemia (diff)", "Monocytes", "Leukemia (Mon)", "Lymphocytes","Leukemia (stem 2)", "NK cells", "Unknown")
names(new.cluster.ids) <- levels(Pat1_CITE.only)
Pat1_CITE.only <- RenameIdents(Pat1_CITE.only, new.cluster.ids)prop.table(table(Pat1_CITE.only@meta.data$Sample, Pat1_CITE.only@meta.data$celltype), margin = 1)*100
Leukemia cells (HSPC-alike)
Pat1_R1_H1-N 20.2292264
Pat1_R1_Ven_H2-N 24.8518957
Leukemia cells (Monocytic progenitor-alike)
Pat1_R1_H1-N 25.7306590
Pat1_R1_Ven_H2-N 21.7120853
Leukemia cells (stem cells)
Pat1_R1_H1-N 9.8567335
Pat1_R1_Ven_H2-N 16.3803318
Leukemia cells (diffuse – Lymphocytes/Megakaryocyte patterns)
Pat1_R1_H1-N 10.4297994
Pat1_R1_Ven_H2-N 13.1220379
Monocytes (and cDCs) Leukemia cells (Monocytic alike)
Pat1_R1_H1-N 13.4670487 10.2578797
Pat1_R1_Ven_H2-N 9.0639810 7.4348341
Lymphocytes (CD4 T-Cell dominant)
Pat1_R1_H1-N 6.0171920
Pat1_R1_Ven_H2-N 4.5023697
Leukemia cells (stem cells 2) NK cells (and T-cells)
Pat1_R1_H1-N 2.1203438 1.4326648
Pat1_R1_Ven_H2-N 1.0959716 1.2440758
Unknown (cDC dominant)
Pat1_R1_H1-N 0.4584527
Pat1_R1_Ven_H2-N 0.5924171DimPlot(Pat1_CITE.only, reduction = "tsne", group.by="celltype", label = TRUE, label.size=4, pt.size = 0.5) + NoLegend()
DimPlot(Pat1_CITE.only, reduction = "umap", group.by="celltype", label = TRUE, label.size=4, pt.size = 0.5) + NoLegend()
DimPlot(Pat1_CITE.only, reduction = "umap", group.by="celltype", label = TRUE, label.size=4, pt.size = 0.5) + theme(legend.position="bottom")
##CITE Antibodies
VlnPlot(Pat1_CITE.only, assay= "Epi", features = c("CD19","CD3","CD16","CD4","CD11c","CD56-NCAM","CD14","CD8","CD45","CD34","CD15","unmapped" ), ncol = 6, pt.size = 0.1,group.by ="Sample" ) + NoLegend()Warning in FetchData(object = object, vars = features, slot = slot): The
following requested variables were not found: unmapped
FeaturePlot(Pat1_CITE.only, features = c("epi_CD19","CD3","CD16","epi_CD4","CD11c","CD56-NCAM","epi_CD14","CD8","CD45","CD34","CD15","unmapped" ))Warning: Could not find CD3 in the default search locations, found in Epi assay
insteadWarning: Could not find CD16 in the default search locations, found in Epi assay
insteadWarning: Could not find CD11c in the default search locations, found in Epi
assay insteadWarning: Could not find CD56-NCAM in the default search locations, found in Epi
assay insteadWarning: Could not find CD8 in the default search locations, found in Epi assay
insteadWarning: Could not find CD45 in the default search locations, found in Epi assay
insteadWarning: Could not find CD34 in the default search locations, found in Epi assay
insteadWarning: Could not find CD15 in the default search locations, found in Epi assay
insteadWarning in FetchData(object = object, vars = c(dims, "ident", features), : The
following requested variables were not found: unmapped
VlnPlot(Pat1_CITE.only, assay= "Epi", features = c("CD19","CD16","CD56-NCAM","CD8","CD34","unmapped"), ncol = 3, pt.size = 0.1) + NoLegend()Warning in FetchData(object = object, vars = features, slot = slot): The
following requested variables were not found: unmapped
VlnPlot(Pat1_CITE.only, assay= "Epi", features = c("CD3","CD4","CD11c","CD14","CD45","CD15"), ncol = 3, pt.size = 0.1) + NoLegend()
RidgePlot(Pat1_CITE.only, assay = "Epi", features = c("CD3","CD14","CD16","CD4", "CD11c","CD56-NCAM"), ncol = 3) + NoLegend()Picking joint bandwidth of 0.136Picking joint bandwidth of 0.119Picking joint bandwidth of 0.151Picking joint bandwidth of 0.103Picking joint bandwidth of 0.111Picking joint bandwidth of 0.11
RidgePlot(Pat1_CITE.only, assay = "Epi", features = c("CD45","CD15", "CD8","CD34", "CD19"), ncol = 3) + NoLegend()Picking joint bandwidth of 0.103Picking joint bandwidth of 0.0875Picking joint bandwidth of 0.191Picking joint bandwidth of 0.0875Picking joint bandwidth of 0.123
FeatureScatter(Pat1_CITE.only, feature1 = "epi_CD15", feature2 = "epi_CD11c") 
FeatureScatter(Pat1_CITE.only, feature1 = "epi_CD15", feature2 = "epi_CD34") 
FeatureScatter(Pat1_CITE.only, feature1 = "epi_CD34", feature2 = "epi_CD11c")
saveRDS(Pat1_CITE.only, file = "./Pat1_CITE_integrated.rds")